Delivering a protein in its active state into a targeted cell is a major technological hurdle. However, the ability to do so would open up numerous applications in both clinical and basic research settings. We propose to develop cell-targeting systems that fulfill this task, composed of two components: the cell targeting module, and the protein cargo. The targeting modules will be based on the molecular scaffolds of single chain variable fragments (scFvs;~25 kDa), designed ankyrin repeat proteins (DARPins;~15 kDa), and affibodies (~ 9 kDa). Theses scaffolds will undergo optimization using phage display technology in order to acquire the needed attributes. These include tight and specific binding to a cell surface antigen, which is followed by efficient internalization and escape into the cytoplasmic compartment. These properties of the delivery scaffold will allow it to act as a transporter of proteins into cells. To test our technology, we will deliver an engineered version of human deoxycytidine kinase (dCKEN), which is a novel enzyme variant that has been endowed with thymidine kinase activity. We will test the efficiency and selectivity of the delivery scaffolds for their ability to ferry dCKEN into HER2 positive cells. The unique catalytic activity of this engineered enzyme will allow us to confine the activation of thymidine analogs only to cells that have internalized the enzyme. In this way, we would have developed a system that can be used to eradicate HER2 positive cells while not affecting other cells. The challenges to the delivery technology are to discover scaffolds that can be obtained at high yield in E. coli, that bind to the cell-surface marker with low nanomolar affinity, and that undergo efficient internalization and escape from endocytic vesicles into the cytoplasm. The novelty of this work stems from the type of delivery scaffolds used, which are much smaller than conventional monoclonal-based targeting systems. The advantages of using these smaller scaffolds include deeper penetration into solid tumors, reduced non-specific binding due to the absence of an Fc region present in monoclonal antibodies, and the ability for production in E. coli. Moreover, every aspect of the delivery scaffold, from binding affinity via internalization propensity, to escape from vesicles into the cytoplasm will be optimized by coupling phage display with the appropriate selection method. PUBLIC HEALTH RELEVANCE: A major technological hurdle confronting cancer therapeutics is how to take advantage of cancer-cell markers to achieve targeted therapy. This application addresses this need by developing an enzyme delivery technology that transports a unique enzyme into the intracellular compartment of cancer cells. Subsequent administration of an otherwise non-toxic prodrug that is converted to its toxic form by the unique enzyme will result in the elimination of the targeted cancer cells. Importantly, this approach will spare healthy tissue.